PERLRECHARCLASS(1)

📦
version 5.42.3 · 22 versions
, in fedora 43, fedora 44

💖 NAME 💖

perlrecharclass — Perl Regular Expression Character Classes 🎀

(that "(1)" up top just means it lives in the Perl Programmers Reference Guide, iykyk 💅)

🌸 DESCRIPTION 🌸

okay so, the top-level documentation about Perl regular expressions lives in perlre, if you want the whole entire picture. 🫶

this page is specifically about character classes in Perl regular expressions — the syntax and the use of them.

so what even is a character class? it's a way of denoting a set of characters, in such a way that one character of the set gets matched. and bestie, remember this, it's SO important: matching a character class consumes exactly one character in the source string. (the source string is the string the regular expression is matched against, no cap.)

there are three types of character classes in Perl regular expressions:

  • the dot
  • backslash sequences
  • the form enclosed in square brackets

but keep in mind — and this is lowkey a trap — the term "character class" is often used to mean just the bracketed form. and certainly, most Perl documentation does that. 🙈

✨ The dot ✨

the dot (or period), ., is probably the most used, and definitely the most well-known character class. by default, a dot matches any character, except for the newline. 🌈

that default can be changed to add matching the newline, by using the single line modifier:

  • for the entire regular expression, with the /s modifier
  • locally, with (?s)
  • and even globally within the scope of use re '/s'

(the \N backslash sequence, described below, matches any character except newline without regard to the single line modifier. so there's your no-modifier-needed option 💅)

here are some examples:

 "a"  =~  /./       # Match
 "."  =~  /./       # Match
 ""   =~  /./       # No match (dot has to match a character)
 "\n" =~  /./       # No match (dot does not match a newline)
 "\n" =~  /./s      # Match (global 'single line' modifier)
 "\n" =~  /(?s:.)/  # Match (local 'single line' modifier)
 "ab" =~  /^.$/     # No match (dot matches one character)

🦋 Backslash sequences 🦋

a backslash sequence is a sequence of characters, the first one of which is a backslash. Perl ascribes special meaning to many such sequences, and some of these are character classes. that is, they match a single character each, provided that the character belongs to the specific set of characters defined by the sequence. ✨

here's a list of the backslash sequences that are character classes. they're discussed in more detail below. (for the backslash sequences that aren't character classes, see perlrebackslash.)

  • \d — match a decimal digit character.
  • \D — match a non-decimal-digit character.
  • \w — match a "word" character.
  • \W — match a non-"word" character.
  • \s — match a whitespace character.
  • \S — match a non-whitespace character.
  • \h — match a horizontal whitespace character.
  • \H — match a character that isn't horizontal whitespace.
  • \v — match a vertical whitespace character.
  • \V — match a character that isn't vertical whitespace.
  • \N — match a character that isn't a newline.
  • \pP, \p{Prop} — match a character that has the given Unicode property.
  • \PP, \P{Prop} — match a character that doesn't have the Unicode property

💜 \N 💜

\N, available starting in v5.12, like the dot, matches any character that is not a newline. the difference is that \N is not influenced by the single line regular expression modifier (see "The dot" above, bestie).

but heads up — the form \N{...} may mean something completely different:

  • when the {...} is a quantifier, it means to match a non-newline character that many times. for example, \N{3} means to match 3 non-newlines; \N{5,} means to match 5 or more non-newlines.
  • but if {...} is not a legal quantifier, it is presumed to be a named character. see charnames for those.

for example, none of \N{COLON}, \N{4F}, and \N{F4} contain legal quantifiers, so Perl will try to find characters whose names are respectively COLON, 4F, and F4. 😭

🍓 Digits 🍓

\d matches a single character considered to be a decimal digit.

  • if the /a regular expression modifier is in effect, it matches [0-9].
  • otherwise, it matches anything that is matched by \p{Digit}, which includes [0-9].

(an unlikely possible exception: under locale matching rules, the current locale might not have [0-9] matched by \d, and/or might match other characters whose code point is less than 256. the only such locale definitions that are legal would be to match [0-9] plus another set of 10 consecutive digit characters; anything else would be in violation of the C language standard, but Perl doesn't currently assume anything in regard to this. 💀)

so what this means is: unless the /a modifier is in effect, \d not only matches the digits '0' - '9', but also Arabic, Devanagari, and digits from other languages. which may cause some confusion. and some security issues. 😭

some digits that \d matches look like some of the [0-9] ones, but have different values. for example:

  • BENGALI DIGIT FOUR (U+09EA) looks very much like an ASCII DIGIT EIGHT (U+0038)
  • LEPCHA DIGIT SIX (U+1C46) looks very much like an ASCII DIGIT FIVE (U+0035)

an application that is expecting only the ASCII digits might be misled. or if the match is \d+, the matched string might contain a mixture of digits from different writing systems that look like they signify a number different than they actually do. num() in Unicode::UCD can be used to safely calculate the value, returning undef if the input string contains such a mixture. otherwise, for example, a displayed price might be deliberately different than it appears. 💀 (the price was lying, bestie, and not in the fun way)

what \p{Digit} means (and hence \d, except under the /a modifier) is \p{General_Category=Decimal_Number}, or synonymously, \p{General_Category=Digit}. starting with Unicode version 4.1, this is the same set of characters matched by \p{Numeric_Type=Decimal}.

but Unicode also has a different property with a similar name — \p{Numeric_Type=Digit} — which matches a completely different set of characters. these characters are things such as CIRCLED DIGIT ONE or subscripts, or are from writing systems that lack all ten digits.

the design intent is for \d to exactly match the set of characters that can safely be used with "normal" big-endian positional decimal syntax, where, for example 123 means one 'hundred', plus two 'tens', plus three 'ones'. this positional notation does not necessarily apply to characters that match the other type of "digit", \p{Numeric_Type=Digit}, and so \d doesn't match them. periodt. 👑

the Tamil digits (U+0BE6 - U+0BEF) can also legally be used in old-style Tamil numbers, in which they would appear no more than one in a row, separated by characters that mean "times 10", "times 100", etc. (see https://www.unicode.org/notes/tn21.)

any character not matched by \d is matched by \D.

🌈 Word characters 🌈

a \w matches a single alphanumeric character (an alphabetic character, or a decimal digit); or a connecting punctuation character, such as an underscore (_); or a "mark" character (like some sort of accent) that attaches to one of those.

it does not match a whole word, bestie! to match a whole word, use \w+. this isn't the same thing as matching an English word, but in the ASCII range it is the same as a string of Perl-identifier characters.

if the /a modifier is in effect ...

  • \w matches the 63 characters [a-zA-Z0-9_].

otherwise ...

  • for code points above 255 ...
    • \w matches the same as \p{Word} matches in this range. that is, it matches Thai letters, Greek letters, etc. this includes connector punctuation (like the underscore) which connect two words together, or diacritics, such as a COMBINING TILDE and the modifier letters, which are generally used to add auxiliary markings to letters.
  • for code points below 256 ...
    • if locale rules are in effect ...
      • \w matches the platform's native underscore character plus whatever the locale considers to be alphanumeric.
    • if, instead, Unicode rules are in effect ...
      • \w matches exactly what \p{Word} matches.
    • otherwise ...
      • \w matches [a-zA-Z0-9_].

which rules apply are determined as described in "Which character set modifier is in effect?" in perlre.

there are a number of security issues with the full Unicode list of word characters. see https://unicode.org/reports/tr36.

also, for a somewhat finer-grained set of characters that are in programming language identifiers beyond the ASCII range, you may wish to instead use the more customized "Unicode Properties", \p{ID_Start}, \p{ID_Continue}, \p{XID_Start}, and \p{XID_Continue}. see https://unicode.org/reports/tr31.

any character not matched by \w is matched by \W.

🎀 Whitespace 🎀

\s matches any single character considered whitespace.

if the /a modifier is in effect ...

  • in all Perl versions, \s matches the 5 characters [\t\n\f\r ]; that is, the horizontal tab, the newline, the form feed, the carriage return, and the space. starting in Perl v5.18, it also matches the vertical tab, \cK. see note "[1]" below for a discussion of this.

otherwise ...

  • for code points above 255 ...
    • \s matches exactly the code points above 255 shown with an "s" column in the table below.
  • for code points below 256 ...
    • if locale rules are in effect ...
      • \s matches whatever the locale considers to be whitespace.
    • if, instead, Unicode rules are in effect ...
      • \s matches exactly the characters shown with an "s" column in the table below.
    • otherwise ...
      • \s matches [\t\n\f\r ] and, starting in Perl v5.18, the vertical tab, \cK. (see note "[1]" below for a discussion of this.) note that this list doesn't include the non-breaking space.

which rules apply are determined as described in "Which character set modifier is in effect?" in perlre.

any character not matched by \s is matched by \S.

\h matches any character considered horizontal whitespace; this includes the platform's space and tab characters and several others listed in the table below. \H matches any character not considered horizontal whitespace. they use the platform's native character set, and do not consider any locale that may otherwise be in use.

\v matches literally any character that counts as vertical whitespace, and bestie that includes the platform's carriage return and line feed characters (aka newline) plus several other characters, all of them listed in the cute little table down below. \V is its opposite era: it matches any character that is not considered vertical whitespace. Both of them use the platform's native character set, and they do not consider any locale that may otherwise be in use. No cap.

\R matches anything that can be considered a newline under Unicode rules. It can match a multi-character sequence, which is kind of a big deal. It cannot be used inside a bracketed character class — use \v instead (vertical whitespace, iykyk). It uses the platform's native character set, and it does not consider any locale that may otherwise be in use. All the juicy details live in perlrebackslash. 🦋

Note that unlike \s (and \d and \w), \h and \v always match the same characters, without regard to other factors, such as the active locale or whether the source string is in UTF-8 format. They're consistent like that. 💅

You might be thinking "isn't \s just equivalent to [\h\v]?" and honestly? Valid thought. It is true starting in Perl v5.18, but prior to that, the sole difference was that the vertical tab (\cK) was not matched by \s. One tiny little glyph. The audacity. 😭

The following table is a complete listing of characters matched by \s, \h and \v as of Unicode 14.0.

The first column gives the Unicode code point of the character (in hex format), the second column gives the (Unicode) name. The third column indicates by which class(es) the character is matched (assuming no locale is in effect that changes the \s matching).

 0x0009        CHARACTER TABULATION   h s
 0x000a              LINE FEED (LF)    vs
 0x000b             LINE TABULATION    vs  [1]
 0x000c              FORM FEED (FF)    vs
 0x000d        CARRIAGE RETURN (CR)    vs
 0x0020                       SPACE   h s
 0x0085             NEXT LINE (NEL)    vs  [2]
 0x00a0              NO-BREAK SPACE   h s  [2]
 0x1680            OGHAM SPACE MARK   h s
 0x2000                     EN QUAD   h s
 0x2001                     EM QUAD   h s
 0x2002                    EN SPACE   h s
 0x2003                    EM SPACE   h s
 0x2004          THREE-PER-EM SPACE   h s
 0x2005           FOUR-PER-EM SPACE   h s
 0x2006            SIX-PER-EM SPACE   h s
 0x2007                FIGURE SPACE   h s
 0x2008           PUNCTUATION SPACE   h s
 0x2009                  THIN SPACE   h s
 0x200a                  HAIR SPACE   h s
 0x2028              LINE SEPARATOR    vs
 0x2029         PARAGRAPH SEPARATOR    vs
 0x202f       NARROW NO-BREAK SPACE   h s
 0x205f   MEDIUM MATHEMATICAL SPACE   h s
 0x3000           IDEOGRAPHIC SPACE   h s

[1] Prior to Perl v5.18, \s did not match the vertical tab. [^\S\cK] (obscurely) matches what \s traditionally did.

[2] NEXT LINE and NO-BREAK SPACE may or may not match \s depending on the rules in effect. See the beginning of this section. 👀

✨ Unicode Properties ✨

\pP and \p{Prop} are character classes that match characters fitting given Unicode properties. One letter property names can be used in the \pP form, with the property name following the \p — otherwise, braces are required. When you're using braces, there's a single form, which is just the property name enclosed in the braces, and then there's a compound form which looks like \p{name=value}, which means match if the property "name" for the character has that particular "value". So for instance, a match for a number can be written as /\pN/ or as /\p{Number}/, or as /\p{Number=True}/. 🌸

Lowercase letters are matched by the property Lowercase_Letter which has the short form Ll. They need the braces, so they're written as /\p{Ll}/ or /\p{Lowercase_Letter}/, or /\p{General_Category=Lowercase_Letter}/ (the underscores are optional). Careful though — /\pLl/ is valid, but it means something different. It matches a two character string: a letter (Unicode property \pL), followed by a lowercase l. Two characters. Sneaky. 🙈

What a Unicode property matches is never subject to locale rules, and if locale rules are not otherwise in effect, the use of a Unicode property will force the regular expression into using Unicode rules, if it isn't already. Forced. Periodt.

Note that almost all properties are immune to case-insensitive matching. That is, adding a /i regular expression modifier does not change what they match. But there are two sets that are affected:

  • The first set is Uppercase_Letter, Lowercase_Letter, and Titlecase_Letter, all of which match Cased_Letter under /i matching.
  • The second set is Uppercase, Lowercase, and Titlecase, all of which match Cased under /i matching. (The difference between these sets is that some things, such as Roman numerals, come in both upper and lower case, so they are Cased, but aren't considered to be letters, so they aren't Cased_Letters. They're actually Letter_Numbers.) This set also includes its subsets PosixUpper and PosixLower, both of which under /i match PosixAlpha.

For more details on Unicode properties, see "Unicode Character Properties" in perlunicode; for a complete list of possible properties, see "Properties accessible through \p{} and \P{}" in perluniprops, which notes all forms that have /i differences. It is also possible to define your own properties. This is discussed in "User-Defined Character Properties" in perlunicode. Go off, define your own. 👑

Unicode properties are defined (surprise!) only on Unicode code points. Starting in v5.20, when matching against \p and \P, Perl treats non-Unicode code points (those above the legal Unicode maximum of 0x10FFFF) as if they were typical unassigned Unicode code points.

Prior to v5.20, Perl raised a warning and made all matches fail on non-Unicode code points. This could be somewhat surprising:

 chr(0x110000) =~ \p{ASCII_Hex_Digit=True}     # Fails on Perls < v5.20.
 chr(0x110000) =~ \p{ASCII_Hex_Digit=False}    # Also fails on Perls
                                               # < v5.20

Even though these two matches might be thought of as complements, until v5.20 they were so only on Unicode code points. 😭

Starting in perl v5.30, wildcards are allowed in Unicode property values. See "Wildcards in Property Values" in perlunicode. ✨

Examples

 "a"  =~  /\w/      # Match, "a" is a 'word' character.
 "7"  =~  /\w/      # Match, "7" is a 'word' character as well.
 "a"  =~  /\d/      # No match, "a" isn't a digit.
 "7"  =~  /\d/      # Match, "7" is a digit.
 " "  =~  /\s/      # Match, a space is whitespace.
 "a"  =~  /\D/      # Match, "a" is a non-digit.
 "7"  =~  /\D/      # No match, "7" is not a non-digit.
 " "  =~  /\S/      # No match, a space is not non-whitespace.

 " "  =~  /\h/      # Match, space is horizontal whitespace.
 " "  =~  /\v/      # No match, space is not vertical whitespace.
 "\r" =~  /\v/      # Match, a return is vertical whitespace.

 "a"  =~  /\pL/     # Match, "a" is a letter.
 "a"  =~  /\p{Lu}/  # No match, /\p{Lu}/ matches upper case letters.

 "\x{0e0b}" =~ /\p{Thai}/  # Match, \x{0e0b} is the character
                           # 'THAI CHARACTER SO SO', and that's in
                           # Thai Unicode class.
 "a"  =~  /\P{Lao}/ # Match, as "a" is not a Laotian character.

It is worth emphasizing that \d, \w, etc, match single characters, not complete numbers or words. To match a number (that consists of digits), use \d+; to match a word, use \w+. But be aware of the security considerations in doing so, as mentioned above. > Stay safe out there bestie. 🫶

🌈 Bracketed Character Classes 🌈

The third form of character class you can use in Perl regular expressions is the bracketed character class. In its simplest form, it lists the characters that may be matched, surrounded by square brackets, like this: [aeiou]. This matches one of a, e, i, o or u. Like the other character classes, exactly one character is matched.* To match a longer string consisting of characters mentioned in the character class, follow the character class with a quantifier. For instance, [aeiou]+ matches one or more lowercase English vowels. Vowels are having a moment. 💖

Repeating a character in a character class has no effect; it's considered to be in the set only once. The set does not care about your duplicates.

Examples:

 "e"  =~  /[aeiou]/        # Match, as "e" is listed in the class.
 "p"  =~  /[aeiou]/        # No match, "p" is not listed in the class.
 "ae" =~  /^[aeiou]$/      # No match, a character class only matches
                           # a single character.
 "ae" =~  /^[aeiou]+$/     # Match, due to the quantifier.

* There are two exceptions to a bracketed character class matching a single character only. Each requires special handling by Perl to make things work:

  • When the class is to match caselessly under /i matching rules, and a character that is explicitly mentioned inside the class matches a multiple-character sequence caselessly under Unicode rules, the class will also match that sequence. For example, Unicode says that the letter LATIN SMALL LETTER SHARP S should match the sequence ss under /i rules. Thus,
 'ss' =~ /\A\N{LATIN SMALL LETTER SHARP S}\z/i             # Matches
 'ss' =~ /\A[aeioust\N{LATIN SMALL LETTER SHARP S}]\z/i    # Matches

For this to happen, the class must not be inverted (see "Negation") and the character must be explicitly specified, and not be part of a multi-character range (not even as one of its endpoints). ("Character Ranges" will be explained shortly.) Therefore,

 'ss' =~ /\A[\0-\x{ff}]\z/ui       # Doesn't match
 'ss' =~ /\A[\0-\N{LATIN SMALL LETTER SHARP S}]\z/ui   # No match
 'ss' =~ /\A[\xDF-\xDF]\z/ui   # Matches on ASCII platforms, since
                               # \xDF is LATIN SMALL LETTER SHARP S,
                               # and the range is just a single
                               # element
Note that it isn't a good idea to specify these types of ranges anyway. Fr fr, just don't. 💀
  • Some names known to \N{...} refer to a sequence of multiple characters, instead of the usual single character. When one of these is included in the class, the entire sequence is matched. For example,
  "\N{TAMIL LETTER KA}\N{TAMIL VOWEL SIGN AU}"
                              =~ / ^ [\N{TAMIL SYLLABLE KAU}]  $ /x;

So yeah — it matches, because \N{TAMIL SYLLABLE KAU} is a named sequence made up of the two characters we're matching against. 💖 And just like the other case where a bracketed class can match more than one character (and for pretty much the same reasons), the class must not be inverted, and the named sequence cannot appear in a range — not even a range where it's both endpoints. 🙈 If you do it anyway: it's a fatal error when the character class is inside the scope of use re 'strict', or inside an extended (?[...]) class. Otherwise, only the first code point gets used, and a regexp-type warning gets raised. 😭

🎀 Special Characters Inside a Bracketed Character Class 🎀

Okay bestie, here's the tea ☕: most of the characters that are meta characters in regex (the ones carrying Big Special Meaning, like ., *, or () lose all their specialness inside a character class, so you can just use them with no escaping. For example, [()] matches either an opening paren or a closing paren, and those parens inside the class don't group or capture anything. Slay. ✨

But heads up: unless your pattern is evaluated in single-quotish context, variable interpolation happens before the bracketed class gets parsed. Watch:

 $, = "\t| ";
 $x =~ m'[$,]';        # single-quotish: matches '$' or ','
 $x =~ q{[$,]}'        # same
 $x =~ m/[$,]/;        # double-quotish: Because we made an
                       #   assignment to $, above, this now
                       #   matches "\t", "|", or " "

The characters that might carry special meaning inside a character class are: \, ^, -, [ and ] — and we go through each of them below. You can escape them with a backslash, though honestly sometimes that's not even needed, in which case the backslash can just be omitted. 💅

The sequence \b is special inside a bracketed character class. Outside the class, \b is an assertion — it points at a spot that doesn't have either two word characters or two non-word characters on either side. But inside a bracketed character class, \b matches a backspace character. Not the same vibe at all, ik. 😅

The sequences \a, \c, \e, \f, \n, \N{NAME}, \N{U+hex char}, \r, \t, and \x are also special, and they mean exactly the same thing they mean outside a bracketed character class. 🫶

Also: a backslash followed by two or three octal digits counts as an octal number.

A [ is not special inside a character class, unless it's the start of a POSIX character class (see "POSIX Character Classes" below). Normally it doesn't need escaping.

A ] is normally either the end of a POSIX character class (again, see "POSIX Character Classes" below), or it signals the end of the bracketed character class. If you want a ] to actually be in the set of characters, you generally must escape it.

But! 🦋 If the ] is the first character of a bracketed character class (or the second, if the first character is a caret), then it does not mean the end of the class — because bestie, you can't have an empty class — and it's just considered part of the set of characters that can be matched, no escaping required.

Examples:

 "+"   =~ /[+?*]/     #  Match, "+" in a character class is not special.
 "\cH" =~ /[\b]/      #  Match, \b inside in a character class
                      #  is equivalent to a backspace.
 "]"   =~ /[][]/      #  Match, as the character class contains
                      #  both [ and ].
 "[]"  =~ /[[]]/      #  Match, the pattern contains a character class
                      #  containing just [, and the character class is
                      #  followed by a ].

🌸 Bracketed Character Classes and the /xx pattern modifier 🌸

Normally, SPACE and TAB characters have zero special meaning inside a bracketed character class — they just get tossed into the list of characters the class matches. But if the /xx pattern modifier is in effect, they're generally ignored, and you can sprinkle them in to make things more readable. 🎀

That said, you cannot add them in the middle of a single construct:

 / [ \x{10 FFFF} ] /xx  # WRONG!

The SPACE in the middle of that hex constant is illegal. 💀

If you want a literal SPACE character, escape it with a backslash, like:

 /[ a e i o u \  ]/xx

That matches the English vowels plus the SPACE character. 👑

And for clarity: you should already have been using \t to specify a literal tab, and \t is completely unaffected by /xx.

🦋 Character Ranges 🦋

Wanting to match a range of characters is super common, and luckily you don't have to list every single one — you get to use the hyphen (-). 🎉 If, inside a bracketed character class, you have two characters separated by a hyphen, it's treated as if all the characters between those two were in the class. So [0-9] matches any ASCII digit, and [a-m] matches any lowercase letter from the first half of the ASCII alphabet.

Now, the two characters sitting on either side of the hyphen are not necessarily both letters or both digits — literally any character is possible, though ngl it's not advisable. ['-?] does contain a range of characters, but most people will have no idea which characters that even means. 🙈 And on top of that, ranges like this can cause portability problems if the code has to run on a platform using a different character set, like EBCDIC.

If a hyphen in a character class can't syntactically be part of a range — for example because it's the first or last character of the class, or because it immediately follows a range — then the hyphen isn't special, so it's just treated as a character to be matched literally. And if you want a hyphen in your set of characters to be matched and its position in the class means it could be read as part of a range, you must escape that hyphen with a backslash. 💅

Examples:

 [a-z]       #  Matches a character that is a lower case ASCII letter.
 [a-fz]      #  Matches any letter between 'a' and 'f' (inclusive) or
             #  the letter 'z'.
 [-z]        #  Matches either a hyphen ('-') or the letter 'z'.
 [a-f-m]     #  Matches any letter between 'a' and 'f' (inclusive), the
             #  hyphen ('-'), or the letter 'm'.
 ['-?]       #  Matches any of the characters  '()*+,-./0123456789:;<=>?
             #  (But not on an EBCDIC platform).
 [\N{APOSTROPHE}-\N{QUESTION MARK}]
             #  Matches any of the characters  '()*+,-./0123456789:;<=>?
             #  even on an EBCDIC platform.
 [\N{U+27}-\N{U+3F}] # Same. (U+27 is "'", and U+3F is "?")

As those last two examples show, you can get portability to non-ASCII platforms by using the \N{...} form for the range endpoints. ✨ These say the specified range should be interpreted using Unicode values, so [\N{U+27}-\N{U+3F}] means "match \N{U+27}, \N{U+28}, \N{U+29}, ..., \N{U+3D}, \N{U+3E}, and \N{U+3F}" — whatever the native code point versions of those happen to be. These are called "Unicode" ranges. If either end is in the \N{...} form, the whole range is considered Unicode. And a regexp warning gets raised under use re 'strict' if the other endpoint is specified in a non-portable way:

 [\N{U+00}-\x09]    # Warning under re 'strict'; \x09 is non-portable
 [\N{U+00}-\t]      # No warning;

Both of those match the characters \N{U+00} \N{U+01} ... \N{U+08}, \N{U+09} — but the \x09 looks like it could be a mistake, so the warning gets raised (under re 'strict') for it. 😌

Perl also guarantees that the ranges A-Z, a-z, 0-9, and any subranges of those, match what an English-only speaker would expect them to match on any platform. That is, [A-Z] matches the 26 ASCII uppercase letters, [a-z] matches the 26 lowercase letters, and [0-9] matches the 10 digits. Subranges, like [h-k], match correspondingly — in that case just the four letters h, i, j, and k. 🍓 On ASCII platforms this is the natural behavior anyway, since the code points (ordinal values) for h through k are consecutive integers (0x68 through 0x6B). But special handling may be needed on platforms with a non-ASCII native character set. For example, on EBCDIC platforms, the code point for h is 0x88, i is 0x89, j is 0x91, and k is 0x92 — and Perl specially treats [h-k] to exclude the seven code points sitting in that gap: 0x8A through 0x90. This special handling is only invoked when the range is a subrange of one of the ASCII uppercase, lowercase, and digit ranges, AND each end of the range is expressed either as a literal, like A, or as a named character (\N{...}, including the \N{U+...} form). 🫶

EBCDIC Examples:

 [i-j]               #  Matches either "i" or "j"
 [i-\N{LATIN SMALL LETTER J}]  # Same
 [i-\N{U+6A}]        #  Same
 [\N{U+69}-\N{U+6A}] #  Same
 [\x{89}-\x{91}]     #  Matches 0x89 ("i"), 0x8A .. 0x90, 0x91 ("j")
 [i-\x{91}]          #  Same
 [\x{89}-j]          #  Same
 [i-J]               #  Matches, 0x89 ("i") .. 0xC1 ("J"); special
                     #  handling doesn't apply because range is mixed
                     #  case

💜 Negation 💜

You can also flip the whole thing and list the characters you don't want to match. Do it by putting a caret (^) as the first character in the character class. So [^a-z] matches any character that is not a lowercase ASCII letter — which, btw, means it includes more than a million Unicode code points. 😳 The class is then said to be "negated" or "inverted".

This syntax is what makes the caret a special character inside a bracketed character class — but only if it's the first character of the class. So if you want the caret itself to be one of the characters matched, either escape the caret or just don't list it first. Easy. 💅

In inverted bracketed character classes, Perl ignores the Unicode rules that normally say a named sequence, and certain characters, should match a sequence of multiple characters when doing caseless /i matching. Following those rules could lead to some truly confusing situations:

 "ss" =~ /^[^\xDF]+$/ui;   # Matches!

That's supposed to match any sequence of characters that isn't \xDF — nor whatever \xDF matches under /i. The thing is, s isn't \xDF, but Unicode says ss is what \xDF matches under /i. So which one wins? 😵‍💫 Do you fail the match because the string has ss, or accept it because it has an s followed by another s? Perl picked the latter. (See the note in "Bracketed Character Classes" above.)

Examples:

 "e"  =~  /[^aeiou]/   #  No match, the 'e' is listed.
 "x"  =~  /[^aeiou]/   #  Match, as 'x' isn't a lowercase vowel.
 "^"  =~  /[^^]/       #  No match, matches anything that isn't a caret.
 "^"  =~  /[x^]/       #  Match, caret is not special here.

⭐ Backslash Sequences ⭐

okay so the wildest glow-up for bracketed character classes? you can yeet a backslash sequence character class inside one. literally any of them — with the exception of "\N" and "\R" 🙅‍♀️ — and it acts exactly like you'd dumped every character that sequence matches straight into the brackets. it's giving buy-one-get-one energy. for instance, [a-f\d] matches any decimal digit, or any of the lowercase letters between 'a' and 'f' inclusive. periodt. 💅

lowkey a trap tho: "\N" inside a bracketed character class must be of the forms "\N{name}" or "\N{U+hex char}", and NOT the form that matches non-newlines. same reason a dot "." inside a bracketed character class loses its special meaning: it matches nearly anything, which generally isn't what you want to happen. 😭

Examples:

 /[\p{Thai}\d]/     # Matches a character that is either a Thai
                    # character, or a digit.
 /[^\p{Arabic}()]/  # Matches a character that is neither an Arabic
                    # character, nor a parenthesis.

and no, a backslash sequence character class can not form one of the endpoints of a range. so this? not it, bestie:

 /[\p{Thai}-\d]/     # Wrong!

🎀 POSIX Character Classes 🎀

POSIX character classes have the form [:class:], where class is the name, and the [: and :] are the delimiters. they only ever appear inside bracketed character classes, and they're a convenient, descriptive way of listing a whole group of characters. cute AND useful. obsessed. ✨

be careful about the syntax tho:

 # Correct:
 $string =~ /[[:alpha:]]/

 # Incorrect (will warn):
 $string =~ /[:alpha:]/

that second pattern? it's just a character class consisting of a colon, and the letters a, l, p and h. not the slay you were going for. 💀

POSIX character classes can be part of a larger bracketed character class. for example,

 [01[:alpha:]%]

is valid and matches '0', '1', any alphabetic character, and the percent sign. 🫶

Perl recognizes the following POSIX character classes:

  • alpha — Any alphabetical character (e.g., [A-Za-z]).
  • alnum — Any alphanumeric character (e.g., [A-Za-z0-9]).
  • ascii — Any character in the ASCII character set.
  • blank — Any horizontal whitespace character (e.g. space or horizontal tab ("\t")).
  • cntrl — Any control character. See Note [2] below.
  • digit — Any decimal digit (e.g., [0-9]), equivalent to "\d".
  • graph — Any printable character, excluding a space. See Note [3] below.
  • lower — Any lowercase character (e.g., [a-z]).
  • print — Any printable character, including a space. See Note [4] below.
  • punct — Any graphical character excluding "word" characters. Note [5].
  • space — Any whitespace character. "\s" including the vertical tab ("\cK").
  • upper — Any uppercase character (e.g., [A-Z]).
  • word — A Perl extension (e.g., [A-Za-z0-9_]), equivalent to "\w".
  • xdigit — Any hexadecimal digit (e.g., [0-9a-fA-F]). Note [7].

like the Unicode properties, most of the POSIX properties match the same regardless of whether case-insensitive ("/i") matching is in effect or not. the two exceptions are "[:upper:]" and "[:lower:]". under /i, they each match the union of "[:upper:]" and "[:lower:]". 🌸

Most POSIX character classes have two Unicode-style "\p" property counterparts. (they are not official Unicode properties, but Perl extensions derived from official Unicode properties.) the table below shows the relation between POSIX character classes and these counterparts.

One counterpart, in the column labelled "ASCII-range Unicode" in the table, matches only characters in the ASCII character set.

The other counterpart, in the column labelled "Full-range Unicode", matches any appropriate characters in the full Unicode character set. For example, "\p{Alpha}" matches not just the ASCII alphabetic characters, but any character in the entire Unicode character set considered alphabetic. An entry in the column labelled "backslash sequence" is a (short) equivalent.

  • alpha — ASCII-range Unicode: \p{PosixAlpha}; Full-range Unicode: \p{XPosixAlpha}
  • alnum — ASCII-range Unicode: \p{PosixAlnum}; Full-range Unicode: \p{XPosixAlnum}
  • ascii — ASCII-range Unicode: \p{ASCII}
  • blank — ASCII-range Unicode: \p{PosixBlank}; Full-range Unicode: \p{XPosixBlank} or \p{HorizSpace} [1]; backslash sequence: \h [1]
  • cntrl — ASCII-range Unicode: \p{PosixCntrl}; Full-range Unicode: \p{XPosixCntrl} [2]
  • digit — ASCII-range Unicode: \p{PosixDigit}; Full-range Unicode: \p{XPosixDigit}; backslash sequence: \d
  • graph — ASCII-range Unicode: \p{PosixGraph}; Full-range Unicode: \p{XPosixGraph} [3]
  • lower — ASCII-range Unicode: \p{PosixLower}; Full-range Unicode: \p{XPosixLower}
  • print — ASCII-range Unicode: \p{PosixPrint}; Full-range Unicode: \p{XPosixPrint} [4]
  • punct — ASCII-range Unicode: \p{PosixPunct}; Full-range Unicode: \p{XPosixPunct} [5]
  • (no [[:...:]] name on this row) — ASCII-range Unicode: \p{PerlSpace}; Full-range Unicode: \p{XPerlSpace}; backslash sequence: \s [6]
  • space — ASCII-range Unicode: \p{PosixSpace}; Full-range Unicode: \p{XPosixSpace} [6]
  • upper — ASCII-range Unicode: \p{PosixUpper}; Full-range Unicode: \p{XPosixUpper}
  • word — ASCII-range Unicode: \p{PosixWord}; Full-range Unicode: \p{XPosixWord}; backslash sequence: \w
  • xdigit — ASCII-range Unicode: \p{PosixXDigit}; Full-range Unicode: \p{XPosixXDigit} [7]

[1] "\p{Blank}" and "\p{HorizSpace}" are synonyms.

[2] Control characters don't produce output as such, but instead usually control the terminal somehow: for example, newline and backspace are control characters. On ASCII platforms, in the ASCII range, characters whose code points are between 0 and 31 inclusive, plus 127 ("DEL") are control characters; on EBCDIC platforms, their counterparts are control characters.

[3] Any character that is graphical, that is, visible. This class consists of all alphanumeric characters and all punctuation characters.

[4] All printable characters, which is the set of all graphical characters plus those whitespace characters which are not also controls.

[5] "\p{PosixPunct}" and "[[:punct:]]" in the ASCII range match all non-controls, non-alphanumeric, non-space characters:

"[-!"#$%&'()*+,./:;<=>?@[\\\]^_`{|}~]"

(although if a locale is in effect, it could alter the behavior of "[[:punct:]]").

The similarly named property, "\p{Punct}", matches a somewhat different set in the ASCII range, namely

"[-!"#%&'()*,./:;?@[\\\]_{}]"

That is, it is missing the nine characters

"[$+<=>^`|~]"

This is because Unicode splits what POSIX considers to be punctuation into two categories, Punctuation and Symbols.

"\p{XPosixPunct}" and (under Unicode rules) "[[:punct:]]", match what "\p{PosixPunct}" matches in the ASCII range, plus what "\p{Punct}" matches. This is different than strictly matching according to "\p{Punct}". Another way to say it is that if Unicode rules are in effect, "[[:punct:]]" matches all characters that Unicode considers punctuation, plus all ASCII-range characters that Unicode considers symbols. 👑

[6] "\p{XPerlSpace}" and "\p{Space}" match identically starting with Perl v5.18. In earlier versions, these differ only in that in non-locale matching, "\p{XPerlSpace}" did not match the vertical tab, "\cK". Same for the two ASCII-only range forms.

[7] Unlike "[[:digit:]]" which matches digits in many writing systems, such as Thai and Devanagari, there are currently only two sets of hexadecimal digits, and it is unlikely that more will be added. This is because you not only need the ten digits, but also the six "[A-F]" (and "[a-f]") to correspond. That means only the Latin script is suitable for these, and Unicode has only two sets of these, the familiar ASCII set, and the fullwidth forms starting at U+FF10 (FULLWIDTH DIGIT ZERO). 🔥

There are various other synonyms that can be used besides the names listed in the table. For example, "\p{XPosixAlpha}" can be written as "\p{Alpha}". All are listed in "Properties accessible through \p{} and \P{}" in perluniprops.

Both the "\p" counterparts always assume Unicode rules are in effect. On ASCII platforms, this means they assume that the code points from 128 to 255 are Latin-1, and that means that using them under locale rules is unwise unless the locale is guaranteed to be Latin-1 or UTF-8. In contrast, the POSIX character classes are useful under locale rules. They are affected by the actual rules in effect, as follows:

  • If the "/a" modifier is in effect ... each of the POSIX classes matches exactly the same as their ASCII-range counterparts.
  • otherwise ...
    • For code points above 255 ... the POSIX class matches the same as its Full-range counterpart.
    • For code points below 256 ...
      • if locale rules are in effect ... the POSIX class matches according to the locale, except:
        • word — also includes the platform's native underscore character, no matter what the locale is.
        • ascii — on platforms that don't have the POSIX ascii extension, this matches just the platform's native ASCII-range characters.
        • blank — on platforms that don't have the POSIX blank extension, this matches just the platform's native tab and space characters.
      • if, instead, Unicode rules are in effect ... the POSIX class matches the same as the Full-range counterpart.
      • otherwise ... the POSIX class matches the same as the ASCII range counterpart.

Which rules apply are determined as described in "Which character set modifier is in effect?" in perlre. 🦋

🙃 Negation of POSIX character classes 🙃

A Perl extension to the POSIX character class is the ability to negate it. This is done by prefixing the class name with a caret ("^"). Some examples:

  • [[:^digit:]] — ASCII-range Unicode: \P{PosixDigit}; Full-range Unicode: \P{XPosixDigit}; backslash sequence: \D
  • [[:^space:]] — ASCII-range Unicode: \P{PosixSpace}; Full-range Unicode: \P{XPosixSpace}; also \P{PerlSpace} / \P{XPerlSpace}; backslash sequence: \S
  • [[:^word:]] — ASCII-range Unicode: \P{PerlWord}; Full-range Unicode: \P{XPosixWord}; backslash sequence: \W

The backslash sequence can mean either ASCII- or Full-range Unicode, depending on various factors as described in "Which character set modifier is in effect?" in perlre.

✨ [= =] and [. .] ✨

Perl recognizes the POSIX character classes [=class=] and [.class.], but does not (yet?) support them. Any attempt to use either construct raises an exception. 💀

🌈 Examples 🌈

okay bestie, you know what time it is — the receipts. 💖

        /[[:digit:]]/            # Matches a character that is a digit.
        /[01[:lower:]]/          # Matches a character that is either a
                                 # lowercase letter, or '0' or '1'.
        /[[:digit:][:^xdigit:]]/ # Matches a character that can be anything
                                 # except the letters 'a' to 'f' and 'A' to
                                 # 'F'.  This is because the main character
                                 # class is composed of two POSIX character
                                 # classes that are ORed together, one that
                                 # matches any digit, and the other that
                                 # matches anything that isn't a hex digit.
                                 # The OR adds the digits, leaving only the
                                 # letters 'a' to 'f' and 'A' to 'F' excluded.

✨ Extended Bracketed Character Classes ✨

(aka "regex_sets" — yes bestie, it has a government name) 💅

Okay so THIS is the fancy bracketed character class. It's giving ✨main character energy✨ for a reason: it's way more readable, way less error-prone, and — the part that makes me feral — it can do set operations, like intersection. Obsessed. 🫶

Check it:

 /(?[ \p{Thai} & \p{Digit} ])/

That matches every digit character that lives in the Thai script. Slay. 🌸

This feature showed up in Perl 5.18 as experimental, and got officially accepted in 5.36. It's a real one now, no cap.

Also: the rules from use re 'strict' apply to this construct. Keep that in mind, it matters later. 👀

Now let's extend the example:

 /(?[ ( \p{Thai} + \p{Lao} ) & \p{Digit} ])/

This one matches digits that are in either the Thai or Laotian scripts. Two scripts, one vibe check. 🦋

Notice all that white space in the examples? Yeah — this construct always has the /xx modifier turned on inside it. Automatically. You don't even have to ask. ✨

The binary operators, bestie

  • & — intersection
  • + — union
  • | — another name for +, so yeah, it means union too
  • - — subtraction (the result matches the set of those code points matched by the first operand, minus any that are also matched by the second operand)
  • ^ — symmetric difference (the union minus the intersection). It's like an exclusive or: the result is the set of code points matched by either operand, but not both.

The one unary operator

  • ! — complement

Precedence tea, because this matters: all the binary operators left associate, and & has higher precedence than the others — and the others all have <em>equal</em> precedence to each other. The unary operator right associates and has the highest precedence. So basically it follows the normal Perl precedence rules for logical operators. If you want to override any of that default precedence and associativity, use parentheses. Periodt. 💅

The big restriction 😭

Everything is a metacharacter. Everything. So you literally cannot refer to single characters by doing something like this:

 /(?[ a + b ])/ # Syntax error!
Syntax error, bestie. It will NOT work. 🚫

The easiest way to name an individual typable character is to wrap it in brackets:

 /(?[ [a] + [b] ])/

(This is the same thing as [ab], fyi.) You could also have written the equivalent:

 /(?[[ a b ]])/

(And yes, you can totally specify single characters by using \x{...}, \N{...}, etc. Go off. 💖)

That last example is actually showing off how you use this construct to specify an <em>ordinary</em> bracketed character class with no extra set operations at all. Note the white space inside it — that's allowed because /xx is automatically turned on within this construct. Cute, right? 🌸

All the other escapes accepted by normal bracketed character classes are accepted here as well. All of them. 🎀

But — and this is the plot twist — because this construct compiles under use re 'strict', unrecognized escapes that would just generate warnings in normal classes are fatal errors here. Same goes for all other warnings from these class elements, plus some practices that don't even currently warn outside re 'strict'. For example, you cannot say:

 /(?[ [ \xF ] ])/     # Syntax error!
Nope. Fatal. Dead. Gone. 💀

You have to have two hex digits after a braceless \x — use a leading zero to make two. Why so strict? Because the whole point is to lower the incidence of typos quietly making your class not match what you thought it would. Which, honestly, is a serve. 👑

One more lil difference: if a regular bracketed character class contains a \p{} or \P{} and gets matched against a non-Unicode code point, a warning may be raised, because the result is not Unicode-defined. But with this extended form? No such warning will come. Quiet. Peaceful. Zen. 🧚

And the final difference between regular bracketed character classes and these: it is not possible to get these to match a multi-character fold. So this:

 /(?[ [\xDF] ])/iu

does <em>not</em> match the string "ss". That's just how it is, bestie. 🙈

Good news though: you don't have to enclose POSIX class names inside double brackets. So both of these work:

 /(?[ [:word:] - [:lower:] ])/
 /(?[ [[:word:]] - [[:lower:]] ])/

Any contained POSIX character classes — including things like \w and \D — respect the /a (and /aa) modifiers. 💜

⚠️ The regex-compile-time thing ⚠️

Note that (?[ ]) is a regex-compile-time construct. Any attempt to use something that isn't knowable at the time the containing regular expression is compiled is a fatal error. In practice, this means just three limitations:

  1. When compiled within the scope of use locale (or the /l regex modifier), this construct assumes that the execution-time locale will be a UTF-8 one, and the generated pattern always uses Unicode rules. What gets matched or not therefore isn't dependent on the actual runtime locale, so tainting is not enabled. But a locale category warning is raised if the runtime locale turns out to not be UTF-8.
  2. Any user-defined property used must already be defined by the time the regular expression is compiled. (But note — this construct can be used <em>instead</em> of such properties. 👀)
  3. A regular expression that otherwise would compile using /d rules and which uses this construct will instead use /u. So this construct is basically Perl's way of saying you don't want /d rules for the <em>entire</em> regular expression containing it.

White space lore: skipping white space applies only to the interior of this construct. There must not be any space between any of the characters that form the initial (?[, and there also can't be space between the closing ]) characters. No gap. None. 🚫

Interpolation, and the trap 💀

Just like in all regular expressions, you can build the pattern up by including variables that get interpolated at regex compilation time. But — currently, each such sub-component should be an already-compiled extended bracketed character class. Like this:

 my $thai_or_lao = qr/(?[ \p{Thai} + \p{Lao} ])/;
 ...
 qr/(?[ \p{Digit} & $thai_or_lao ])/;

If you interpolate something <em>else</em>? The pattern may still compile (or it may die), but if it compiles, it very well may not behave as you would expect. Not me warning you fr fr. 😭

 my $thai_or_lao = '\p{Thai} + \p{Lao}';
 qr/(?[ \p{Digit} & $thai_or_lao ])/;

compiles to:

 qr/(?[ \p{Digit} & \p{Thai} + \p{Lao} ])/;

And this does not have the effect that someone reading the source code would likely expect, because the intersection applies just to \p{Thai}, excluding the Laotian. So it's not what you meant. It's a jump scare. 🙈

Last thing, I promise: due to the way that Perl parses things, your parentheses and brackets may need to be balanced, <em>even including comments</em>. Yeah. Even comments. If you run into any examples of this, please submit them to <https://github.com/Perl/perl5/issues>, so we can have a concrete example for this man page. Do it for the culture. 🫶

And that's the whole page, bestie!! 🎀 This was perl v5.42.3, dated 2026-08-03, filed under PERLRECHARCLASS(1). Go forth and slay your character classes. Byeeee 💖✨🦋

📦 where this page comes from